Clinical impact of mild acute kidney injury after cardiac surgery.

BACKGROUND

Dialysis-dependent renal failure occurs infrequently after cardiac surgery but leads to substantial morbidity and mortality. In contrast, milder degrees of acute kidney injury (AKI), based on small increases in serum creatinine, occur frequently but the independent impact of mild AKI on outcome remains unclear.

METHODS

Between January 2010 and December 2012, 3,869 consecutive patients undergoing cardiac surgery comprised the study cohort. Acute kidney injury was defined according to the AKI Network criteria as stage I, II, or III. A nonparsimonious multivariable logistic regression model including preoperative and intraoperative variables was constructed to determine a propensity score for the development of stage I AKI followed by a greedy matching algorithm to create 1:1 propensity-matched pairs.

RESULTS

The incidence of stage I AKI in the entire cohort was 22.4%. Stage I AKI patients were more likely to be older; to have diabetes mellitus, hypertension, preoperative renal dysfunction, and poorer left ventricle function; and to require more urgent surgery and longer cardiopulmonary bypass. After propensity matching, the 833 matched pairs were similar in terms of all of the above characteristics (all p > 0.5). Within the matched cohort, AKI patients had higher mortality (2.6% versus 1.2%, p = 0.01), higher incidence of neurologic dysfunction (15.2% versus 8.1%, p < 0.001), and longer duration of mechanical ventilation (41.7 ± 125.0 versus 19.3 ± 58.6 hours, p < 0.001). Intensive care unit stay (5.2 ± 10.7 versus 2.7 ± 3.8 days, p < 0.0001), and hospital length of stay (17.9 ± 20.1 versus 14.7 ± 18.3 days, p = 0.0007) was significantly longer for matched AKI patients.

CONCLUSIONS

Patients with even mild degrees of AKI have increased mortality and morbidity compared with their matched counterparts. Interventions that prevent or mitigate AKI after cardiac surgery can yield substantial clinical benefit.